[Stackless] Mailing list is moved
tismer at tismer.com
Sat Jan 19 01:15:54 CET 2002
Sam Rushing wrote:
> Christian Tismer wrote:
>> I had a look at your coroutine package again, to see how
>> they modify the stack. I thought they copy something, but
>> it appears that only the stack pointer is changed. That's
>> not my target, it consumes too much memory, and it doesn't
>> help on the Palm.
> Yeah, all we're doing is swapping tstate->frame out, managing our own
> coroutine objects.
> But we have a lot of other code written in continuation-passing-style -
> pushing frames with continuation pointers (like map_nr). Will this
> still be possible with the new scheme?
At first glance, not. It depends how we model these frame
chains which are waiting to run. Of course they are like
continuations. Without further actions, they are only
restartable once. I don't want to support full continuations
again. But passing handles to wrapper objects for these chains
should of course be possible.
I want to restrict things a little. Since I'm handling real
stack slices, it must be ensured that a chain can never be
activated in the wron real thread. Therefor, I want to tie
all chains of a real thread to its tstate object and provide
no way to break this rule.
Must be careful with this stuff...
ciao - chris
Christian Tismer :^) <mailto:tismer at tismer.com>
Mission Impossible 5oftware : Have a break! Take a ride on Python's
Kaunstr. 26 : *Starship* http://starship.python.net/
14163 Berlin : PGP key -> http://wwwkeys.pgp.net/
PGP Fingerprint E182 71C7 1A9D 66E9 9D15 D3CC D4D7 93E2 1FAE F6DF
where do you want to jump today? http://www.stackless.com/
Stackless mailing list
Stackless at www.tismer.com
More information about the Stackless